Designing a wireless application today has boiled down to selecting the appropriate chips and integrating them into the product. Often the embedded controller and related software dominate the design process. But one other factor is absolutely crucial to the success of the design-the antenna.
First-time wireless designers often underestimate the importance of this passive collection of conductors, which are a must for good performance. As it turns out, the antenna is the one area where the designer has the most influence. Using the best antenna will enable those chips to deliver their maximum potential range and data rate. In many designs, the antenna can be made right on the pc board holding the circuitry. This copper pattern also can be optimized for the design.
The basic half-wavelength (µ/2) center-fed dipole and quarter-wavelength (µ/4) monopole or ground plane are still widely used (see "Remembering Wavelength," ED Online 9985). However, numerous variations have been created to improve performance or adapt to limited physical conditions, some of which are more common in wireless products. Some of these antennas can be made right on the radio pc board. The most popular are the loop, patch, inverted-F, and meander line.
The Loop The simplest type of antenna, the loop, is just a closed loop of pc-board copper connected to the transmitter's or receiver's antenna terminals (see the figure, a). It may be a round or rectangular loop, but designers should make its circumference as great as possible. The larger the loop, the more efficient it is, and the better it works.
Circumferences of less than one wavelength (µ) are okay, but anything less than 0.1µ is virtually useless. One possible variation is the µ/2 loop that's open-circuited at the top. In this way, it looks and acts like a dipole. Designers feed it as if it were a dipole.
To use a closed loop, connect a capacitor in parallel with it to resonate at the operating frequency. The radiation resistance of this combination is only about 10 V at 0.5µ. This is a difficult match for most transmitters and receivers whose standard impedance is 50 V. Some kind of impedance matching network is essential. If you can get the full µ loop, its resistance is about 120 V, and this can be more readily matched. Overall, the loop is very inefficient, but it works fine in very short-range applications like garage-door and remote keyless entry systems.
The Patch A rectangular or circular copper area on the pc board, the patch antenna spans about one half-wavelength in diameter (see the figure, b). For a rectangular patch, the width is closer to µ/2 divided by the square root of the dielectric constant of the pc board (ve).The patch is backed up on a ground plane on the other side of the pc board. It's usually fed with a microstrip line at the center edge of the patch. Impedance is in the 120-V range; impedance matching with a microstrip line works well. Alternately, the patch can be fed with 50-V coax by connecting the center conductor to the patch off center and shield to the ground plane.
The patch works rather well. It has a slight gain and strong radiation in a direction perpendicular to the patch. This antenna is used in some 802.11 wide local-area-network (WLAN) antennas, as well as in phased arrays, to increase gain and narrow beam width. Bandwidth is narrow, but it can be increased by using a thicker dielectric material, thereby increasing the spacing between the patch and the ground plane.
The Inverted-F Another widely used pc-board antenna is the inverted-F (see the figure, c). It's found in cell phones, WLAN hardware, and other small wireless devices. The performance is similar to a quarter-wave ground plane. Note that the dimensions in the figure are best determined experimentally to get the best impedance match.A large ground plane is necessary to make the antenna efficient, but the radiation pattern is essentially omnidirectional. A planar version of the inverted-F is also popular. (The long part of the F is a plane rather than a copper strip.) It greatly expands the antenna bandwidth. This is very desirable given the very broadband nature of digital wireless today.
The Meander Line Another antenna variant is the meander line (see the figure, d). This can be in a µ/2 dipole or µ/4 ground plane format. The idea is to fold the conductors back and forth to make the overall antenna shorter. The result is a smaller area, but the radiation resistance, efficiency, and bandwidth decrease, too. Experimentation is needed to find the right combination, and some form of impedance matching is typically necessary.The Slot Finally, there's the slot antenna (see the figure, e). It's like the photographic negative of a patch antenna, as it's nothing more than a µ/2 long slot cut in a sheet of metal or the copper plane on a pc board. It acts like a µ/2 dipole, except that its electric and magnetic fields are opposite those of a dipole.A horizontally oriented slot has vertical electric field polarization. The slot is fed at the center, and the impedance is relatively high (usually several hundred ohms). Therefore, some matching is required. The slot is widely used in aircraft radars and in phased arrays.
Resources:
1. Freescale (formerly Motorola) Semiconductor Application Note AN2731/D, Compact, Integrated Antennas: Designs and Applications for the MC13191 and MC13192.
2. Micrel Inc. Application Note 23, MICRF001 Antenna Design Tutorial
3. Kai Chang, RF and Microwave Wireless Systems, John Wiley & Sons Inc., 2000.
4. Alan Bensky, Short-Range Wireless Communication, LLH Technology Publishing/Elsevier, 2000.